PhysicsProf Steven E. Jones circuit shows 8x overunity ?

[JouleSeeker]

@ JouleSeeker

I think your circuit has a problem. I replaced the green LED with a 10mm white high power LED and my circuit still draws 4.0 uA.

Please clarify -- which circuit? and what is the problem you see?

[JouleSeeker]

  @Nick -- I ended the AA-battery test on the sj1 circuit after 2 days (+13 minutes).  The voltage was 1.617V, beginning V was 1.623V.

   Returning to the sj1 circuit, I found that the red diode dims down to 0.89V, then becomes suddenly brighter again.  This is similar to what I observed (as noted above) with my build of Xee2's circuit.

So, I remeasured the Pinput for this region FOLLOWING the re-brightening of the diode in the sj1 circuit (with 50 kohms to the base of the transistor), with the result that the power consumption dropped dramatically, to 8.5uW.

So again, I see evidence that this is an important effect, this re-brightening of the diode (now in both circuits) at a critical voltage.  For below V-crit, the power consumption drops off while the brightness of the LED increases.  I'm enthused by these results -- its an intriguing effect IMO, one that I'd like to better understand.

Next, by increasing the R to the base from 50 Kohms to 390Kohms, the power consumption goes down to  4.6uW, and the brightening occurs at 1.03V.
(0.90 V to 0.86 V on a 10mF cap, in 77 s).

[xee2]

Please clarify -- which circuit? and what is the problem you see?

You were getting changes in input power for different LEDs in your copy of my circuit. I do not. The input power is primarily set by the base current. The LED used should not effect this.

[NickZ]

   @ JS:
    You have some great results, but, if the led is barely lit, what purpose does it serve?  With Koolers devices the leds are bright, even three of them for months.
   I do have one BwJt that if I turn the pot down all the way the led is about 1/2 brightness and after two days still shows almost the same voltage, but the current has dropped to half of what it started at.  Still this show that I'm getting closer to your results, even without using the Goldmine toroids, which I think are the key to your success. 
  The point is that you are getting there by sacrificing the brightness of the led light.  Now maybe try to maintain the brightness from getting dim, and still maintain the same voltage and current. That is where using or tuning to the sweet spot will help. 
   Thanks again for doing your AA test.  I do think that it shows more than your cap test, as the cap test only show discharging of the cap, where as the AA show no sign of loss, or hardly any.
     

[nul-points]

 
Steven

here's an update on the logging results for my SJ1 variant with interrupted supply

as you can see, it does look like it was a case of 'battery relaxation', as expected

i'm going to leave the test running for a while longer, though, because now that the peak cell voltage has been passed, i can observe the comparative effects of different config changes on the cell discharge slope

i've dug out my 4-channel USB micro-logger and connected it up to my  latest SJ1 variant - i've been trying out a few different ideas with either NPN or PNP configs (battery supplied) and i can log a single cell terminal voltage, and coil/core and ambient temperatures simultaneously for sustained lengths of time


i'm following your investigation of the 'critical' point with interest - when you first mentioned it, i thought you were referring to the slight 'burst' of light which sometimes occurs immediately before a capacitor supplied LED circuit finally extinguishes - which i put down to some non-linearity in the transistor transfer function around a volt or so (a more 'active' region which suddenly 'consumed' the remaining usable energy - and where there is a slight current surge associated)

but obviously, the effect which you've noticed is indeed related to different levels of efficiency, apparently within the LEDs themselves - a good catch!

greetings
np


http://docsfreelunch.blogspot.com